Using advanced artificial intelligence can give patients a personalized assessment of their risk for significant events, such as heart attack or death.
Deep learning does a better job in predicting which patients will experience major adverse cardiac events than the current standard imaging protocols used to make these determinations, according to new research.
In a poster presented during the Society of Nuclear Medicine and Molecular Imaging 2021 Annual Meeting, researchers from Cedars-Sinai Medical Center in Los Angeles created a novel deep learning algorithm that can give patients a personalized assessment of their annualized risk for these types of events, such as heart attack of death.
For more SNMMI 2021 Annual Meeting coverage, click here.
“These findings show that artificial intelligence could be incorporated in standard clinical workstations to assist physicians in accurate and fast risk assessment of patients undergoing SPECT MPI scans,” said Ananya Singh, MS, a research software engineer in the Slomka Lab at Cedars-Sinai. “This work signifies the potential advantage of incorporating artificial intelligence techniques in standard imaging protocols to assist readers with risk stratification.”
Prediction performance of DL compared to quantitative measures and Kaplan-Meier curves for quartiles of DL.
Credit: Image created by Singh et al., Cedars-Sinai Medical Center, Los Angeles, CA.
To build their deep learning network, the team pulled data from 20,401 patients included in the largest available multi-center SPECT dataset – REgistry of Fast myocardial perfusion Imaging in NExt generation SPECT (REFINE SPECT). All patients had undergone SPECT MPI. The team used their network to score each patient on their individual likelihood to have a major adverse cardiac event during the average follow-up period of 4.7 years.
According to their analysis, the network was able to pinpoint the areas of the heart that were associated with an elevated risk of these events and return a risk score in less than 1 second. They determined that the patients with the highest deep learning scores had a 9.7-percent annual major cardiac event rate – a 10.2-fold increased risk over patients with the lowest scores.
For more coverage based on industry expert insights and research, subscribe to the Diagnostic Imaging e-Newsletter here.
Study with CT Data Suggests Women with PE Have More Than Triple the One-Year Mortality Rate than Men
April 3rd 2025After a multivariable assessment including age and comorbidities, women with pulmonary embolism (PE) had a 48 percent higher risk of one-year mortality than men with PE, according to a new study involving over 33,000 patients.
The Reading Room: Racial and Ethnic Minorities, Cancer Screenings, and COVID-19
November 3rd 2020In this podcast episode, Dr. Shalom Kalnicki, from Montefiore and Albert Einstein College of Medicine, discusses the disparities minority patients face with cancer screenings and what can be done to increase access during the pandemic.
Predicting Diabetes on CT Scans: What New Research Reveals with Pancreatic Imaging Biomarkers
March 25th 2025Attenuation-based biomarkers on computed tomography (CT) scans demonstrated a 93 percent interclass correlation coefficient (ICC) agreement across three pancreatic segmentation algorithms for predicting diabetes, according to a study involving over 9,700 patients.
Can Photon-Counting CT be an Alternative to MRI for Assessing Liver Fat Fraction?
March 21st 2025Photon-counting CT fat fraction evaluation offered a maximum sensitivity of 81 percent for detecting steatosis and had a 91 percent ICC agreement with MRI proton density fat fraction assessment, according to new prospective research.